HIV-infected children are often given highly active anti-retroviral therapy (HAART) to reduce HIV-associated disease. The long-term effects and toxicities associated with this chronic therapy in children are not known, but severe cardiotoxicity has been suggested in animal models. This study will use the NIH-sponsored WITS and P2C2 HIV-infected pediatric cohorts to determine how left ventricular (LV) function (particularly fractional shortening and contractility) and structure (particularly wall thickness and mass), are affected by cumulative intensity of HAART exposure. The P2C2 HIV-infected pediatric cohort received non-HAART therapies in various intensities. Yet, this cohort has exhibited persistent and significant depression of LV contractility compared to uninfected children after 5 years of follow-up. These same echocardiographic measures have proven to be independently predictive of mortality. Most of the children in the WITS HIV-infected pediatric cohort have been exposed to HAART at varying times and at varying regimen intensities. By assessing LV structure and function with the same echocardiographic protocol in the WITS cohort as was used previously in the P2C2 cohort, we will be able to determine the incremental effects of HAART and non-HAART therapies on LV structure and function. In addition, the hypothesis that HAART results in impaired mitochondrial function resulting in cardiomyopathy will be assessed by comparison of the parameters of LV structure and function that define cardiomyopathy to the frequency of mitochondrial DNA mutations in cells from these same patients. This will be done through a nested-case-control study of mitochondrial mutations to assess the relationship between HAART, mitochondrial compromise and LV structure and function. Treatment intensity for both HAART and non-HAART regimens will be captured through a cumulative score based on an existing 8-point ordinal scale. Intensity will be measured at 3 points in time: in utero; during the first year of life; and after the first year of life. Analysis of the longitudinal echocardiographic and mitochondrial data will provide valuable information about dose intensity and the comparative impact of HAART versus less aggressive drug regimens, and about the impact of therapy during different stages of child development. Similar longitudinal data on viral load and duration of HIV will enable us to control for the effects of HIV infection on cardiovascular toxicity. The findings will help determine the need for cardiovascular follow-up, prevention and therapeutic trials.